Trellis precoding for the broadcast channel

This paper considers the vector Gaussian broadcast channel where a single transmitter with multiple antennas sends independent information to multiple receivers. An achievable rate region is derived by decomposing the broadcast channel into a series of single-user channels with non-causal side information. The side information may be completely pre-subtracted using precoding techniques. A practical trellis precoding method is presented. Trellis precoding can be viewed as a generalization of the Tomlinson-Harashima(1971, 1969) precoder. By taking into account the entire non-causal side-information sequence, a trellis precoder gives an additional shaping gain up to 1.53 dB compared to a Tomlinson precoder.

[1]  Thomas M. Cover,et al.  Broadcast channels , 1972, IEEE Trans. Inf. Theory.

[2]  T. Cover,et al.  Writing on colored paper , 2001, Proceedings. 2001 IEEE International Symposium on Information Theory (IEEE Cat. No.01CH37252).

[3]  Giuseppe Caire,et al.  On achievable rates in a multi-antenna broadcast downlink , 2000 .

[4]  Patrick P. Bergmans,et al.  A simple converse for broadcast channels with additive white Gaussian noise (Corresp.) , 1974, IEEE Trans. Inf. Theory.

[5]  Wei Yu,et al.  Sum capacity of a Gaussian vector broadcast channel , 2002, Proceedings IEEE International Symposium on Information Theory,.

[6]  G. David Forney,et al.  Trellis precoding: Combined coding, precoding and shaping for intersymbol interference channels , 1992, IEEE Trans. Inf. Theory.

[7]  Max H. M. Costa,et al.  Writing on dirty paper , 1983, IEEE Trans. Inf. Theory.

[8]  Katalin Marton,et al.  A coding theorem for the discrete memoryless broadcast channel , 1979, IEEE Trans. Inf. Theory.

[9]  M. Tomlinson.  New automatic equaliser employing modulo arithmetic , 1971 .

[10]  G. Forney,et al.  Trellis shaping , 1992, IEEE/CAM Information Theory Workshop at Cornell.

[11]  Andrea J. Goldsmith,et al.  The capacity region of broadcast channels with intersymbol interference and colored Gaussian noise , 2001, IEEE Trans. Inf. Theory.